registration-autofuse

data-driven fusion strategy for deformable image registration

An Automatic Fusion network (AutoFuse) that provides flexibility to fuse information at many potential network locations. A Fusion Gate (FG) module is used to control how to fuse information at each potential location based on training data. AutoFuse can automatically optimize its fusion strategy during training and can be generalizable to both unsupervised registration (without any labels) and semi-supervised registration (with weak labels provided for partial training data).
Citations: see the repo
Documentation: https://github.com/mungomeng/registration-autofuse
Source: https://github.com/mungomeng/registration-autofuse
Jupyter: available as a jupyter kernel on https://max-jhub.desy.de
Maxwell: module load maxwell mdlma/unified-framework